Z. Ning

collaborators

4 papers

cs.CL2026

SLAI T-Rex: Full-Parameter Post-training of the DeepSeek-V4 Family on Ascend SuperPOD

Dongfang Li, Xiaodong Luo, Ruoyu Sun +62

Full-parameter post-training of trillion-parameter-scale MoE models introduces substantial system-level challenges for large-scale distributed training, including severe memory pre…

hep-ex2026

Observation of $D_{s}^{+} \to a_{0} (980)^{+} f_{0} (500)$ in the Amplitude Analysis of $D_{s}^{+} \to π^{+} π^{0} π^{0} η$

BESIII Collaboration, M. Ablikim, M. N. Achasov +716

We report the first observation of the decay $D_{s}^{+} \to π^{+} π^{0} π^{0} η$ in a data set corresponding to an integrated luminosity of 7.33 fb $^{- 1}$ , collected in $e^{+} e^{-}$ collisions by t…

physics.ins-det2026

Real-Time Wiener Deconvolution for feature reconstruction in JUNO

L. Lastrucci, M. Grassi, A. Triossi +74

In particle physics, experiments generate substantial amounts of data that can be difficult to process without preliminary scaling. To avoid losing potentially crucial data, experi…

cs.DC2025

High-Throughput LLM inference on Heterogeneous Clusters

Yi Xiong, Jinqi Huang, Wenjie Huang +6

Nowadays, many companies possess various types of AI accelerators, forming heterogeneous clusters. Efficiently leveraging these clusters for high-throughput large language model (L…

SLAI T-Rex: Full-Parameter Post-training of the DeepSeek-V4 Family on Ascend SuperPOD

Observation of Ds+​→a0​(980)+f0​(500) in the Amplitude Analysis of Ds+​→π+π0π0η

Real-Time Wiener Deconvolution for feature reconstruction in JUNO

High-Throughput LLM inference on Heterogeneous Clusters

Observation of $D_{s}^{+} \to a_{0} (980)^{+} f_{0} (500)$ in the Amplitude Analysis of $D_{s}^{+} \to π^{+} π^{0} π^{0} η$